Pharmacokinetics and Dialytic Clearance of Ceftazidime-Avibactam in a Critically Ill Patient on Continuous Venovenous Hemofiltration

ABSTRACT

Ceftazidime-avibactam administered at 1.25 g every 8 h was used to treat multidrug-resistant Pseudomonas aeruginosa bacteremia in a critically ill patient on continuous venovenous hemofiltration (CVVH). Prefiltration plasma drug concentrations of ceftazidime and avibactam were measured at 0, 1, 2, 4, 6, and 8 h along with postfiltration and ultrafiltrate concentrations at h 2 and h 6. Plasma pharmacokinetic parameters of ceftazidime and avibactam, respectively, were as follows: maximum plasma concentration (C_max), 61.10 and 14.54 mg/liter; minimum plasma concentration (C_min), 31.96 and 8.45 mg/liter; half-life (t_1/2), 6.07 and 6.78 h; apparent volume of distribution at the steady state (V_ss), 27.23 and 30.81 liters; total clearance at the steady state (CL_ss), 2.87 and 2.95 liters/h; area under the concentration-time curve from 0 to 8 h (AUC_0–8), 347.87 and 85.69 mg · h/liter. Concentrations of ceftazidime in plasma exceeded the ceftazidime-avibactam MIC (6 mg/liter) throughout the 8-h dosing interval. Mean CVVH extraction ratios for ceftazidime and avibactam were 14.44% and 11.53%, respectively, and mean sieving coefficients were 0.96 and 0.93, respectively. The calculated mean clearance of ceftazidime by CVVH was 1.64 liters/h and for avibactam was 1.59 liters/h, representing 57.1% of the total clearance of ceftazidime and 54.3% of the total clearance of avibactam. Further data that include multiple patients and dialysis modes are needed to verify the optimal ceftazidime-avibactam dosing strategy during critical illness and CVVH.

INTRODUCTION

Ceftazidime-avibactam is a novel oxyimino cephalosporin-diazobicyclooctane β-lactamase inhibitor combination antimicrobial approved for the treatment of complicated urinary tract and intra-abdominal infections (1, 2). Avibactam is capable of inhibiting β-lactamase enzymes from Ambler classes A and C, including KPC and chromosomal Pseudomonas AmpC enzymes, and displays potent in vitro activity against a variety of multidrug-resistant Gram-negative pathogens when combined with ceftazidime (3). This broad in vitro activity and familiarity with ceftazidime have stimulated the rapid uptake and frequent off-label use of ceftazidime-avibactam (4–6). Despite this pervasive use, there are no published pharmacokinetic data to guide dosing in critically ill patients or during the use of any form of continuous renal replacement therapy. Furthermore, the pharmacokinetics and the optimal dosing scheme of ceftazidime-avibactam were the subjects of significant controversy during the regulatory process (3). Therefore, real-world data are needed to guide appropriate dosing in these critically ill patients with multidrug-resistant infections and significant pathophysiological and pharmacokinetic alterations. This work details the plasma pharmacokinetics and dialytic clearance of ceftazidime-avibactam in a patient on continuous venovenous hemofiltration (CVVH) being treated for multidrug-resistant P. aeruginosa bacteremia.

A 53-year-old female patient with a history of a deceased-donor renal transplant was admitted to the medical intensive care unit for idiopathic liver failure. Her hospital course was complicated by toxic megacolon secondary to Clostridium difficile infection, polymicrobial empyema, and anuric renal failure necessitating the initiation of renal replacement therapy. She developed septic shock approximately 3 weeks into her hospitalization, and ascitic fluid, respiratory, and blood cultures were positive for multidrug-resistant P. aeruginosa. The antibiotic susceptibilities for the isolate cultured from ascitic fluid are displayed in Table 1. Susceptibilities to ceftazidime-avibactam and colistin were determined by Etest (bioMérieux, Durham, NC), to ceftolozane-tazobactam by disk diffusion (Hardy Diagnostics, Santa Maria, CA), and to the remaining antimicrobials by Vitek II (bioMérieux, Durham, NC). Given the resistance profile, the patient was started on ceftazidime-avibactam (1.25 g every 8 h) as a 2-h infusion while on CVVH via a NxStage System One machine (NxStage Medical, Inc., Lawrence, MA) with a 1.6-m² polyethersulfone membrane filter. The blood and ultrafiltration flow rates were fixed at 200 ml/min and 2 liters/h, respectively, and replacement fluid was added prefiltration. There were no interruptions in CVVH during her treatment with ceftazidime-avibactam, and the CVVH filter was changed 4 h prior to administration of dose 4.

TABLE 1.

Pseudomonas aeruginosa antibiotic susceptibilities

Antibiotic	MIC (mg/liter) or disk diffusion diam of inhibition	Interpretationa
Amikacin	16	Susceptible
Aztroenam	>16	Resistant
Cefepime	>16	Resistant
Ceftazidime	>16	Resistant
Ceftazidime-avibactam	6/4	Susceptibleb
Ceftolozane-tazobactam	24c	Susceptible
Ciprofloxacin	>2	Resistant
Colistin	2	Susceptible
Gentamicin	8	Intermediate
Imipenem-cilastatin	>8	Resistant
Levofloxacin	>4	Resistant
Meropenem	>8	Resistant
Pierpacillin-tazobactam	>64	Resistant
Ticarcillin-clavulanate	>64	Resistant
Tobramycin	2	Susceptible

^aData were interpreted according to CLSI M100-S27 except where otherwise indicated.

^bData were interpreted using FDA breakpoints as follows: susceptible, ≤8/4 mg/liter; resistant, ≥16/4 mg/liter.

^cDisk diffusion diameter of inhibition.

RESULTS

The prefiltration plasma pharmacokinetic parameters of ceftazidime and avibactam from the patient receiving CVVH and reference parameters from healthy subjects receiving the same dose are displayed in Table 2 (7). Prefiltration plasma concentration-time profiles of ceftazidime and avibactam are displayed in relation to the MIC in Fig. 1. Concentrations of ceftazidime in plasma exceeded the ceftazidime-avibactam MIC throughout the 8-h dosing interval. The extraction ratios for ceftazidime at 2 and 6 h after the start of the infusion were 13.57% and 15.30% and for avibactam were 10.88% and 12.18%, respectively. Sieving coefficients (SC) at the same time points for ceftazidime were 0.88 and 1.03 and for avibactam were 0.89 and 0.97, respectively. The calculated CL_CVVH values for ceftazidime at 2 and 6 h were 1.51 liters/h and 1.77 liters/h and for avibactam were 1.52 liters/h and 1.65 liters/h. The CL_CVVH thus represented 57.1% of the total clearance of ceftazidime and 54.3% of the total clearance of avibactam.

TABLE 2.

Plasma pharmacokinetic parameters of ceftazidime and avibactam in a critically ill patient receiving CVVH and in healthy subjects

Parameter	Value
	Ceftazidime		Avibactam
	CVVH patienta	Healthy volunteersb	CVVH patienta	Healthy volunteersb
Cmax (mg/liter)	61.10	57.35	14.54	12.93
Cmin (mg/liter)	31.96		8.45
t1/2 (h)	6.07	1.82	6.78	1.79
AUC (mg · h/liter)	347.87c	128.26d	85.69c	20.66d
CL (liters/h)	2.87	7.74	2.92	12.03
Vss (liters)	27.23	18.88	30.81	21.08

^aCVVH settings: filter, 1.6-m² polyethersulfone membrane; blood flow rate, 200 ml/min; prefilter replacement fluid rate, 2 liters/h.

^bHealthy volunteers were given a single dose of 1,000 mg of ceftazidime and 250 mg avibactam (8). Data represent mean or median values.

^cAUC_0–8 (AUC from 0 to 8 hours).

^dAUC_0–last.

FIG 1.

Prefiltration concentration-time profiles of ceftazidime (solid line, filled circles) and avibactam (dotted line, open circles) in plasma before and after the fourth dose of ceftazidime-avibactam (1.25 g) in relation to the MIC of the infecting P. aeruginosa pathogen (dashed horizontal line). The shaded region represents the infusion period, accounting for overfill volume and residual fluid volume in the tubing. Data in the y axis are in the log scale.

DISCUSSION

To our knowledge, this is the first published report of a patient receiving ceftazidime-avibactam while on any type of renal replacement therapy. Given the lack of available data and considering the severity of the illness of the patient and the MIC of the pathogen, the dose of ceftazidime-avibactam used in this case was extrapolated from the package insert dosing recommendations for patients with an estimated creatinine clearance of 31 to 50 ml/min. We then prospectively measured the systemic concentrations and dialytic clearance of ceftazidime-avibactam in order to calculate patient-specific pharmacokinetic parameters and optimize dosing.

Our report demonstrates significant pharmacokinetic alterations during critical illness and renal replacement therapy compared to healthy volunteers administered the same dose of ceftazidime-avibactam (Table 2). The plasma exposure in terms of area under the concentration-time curve (AUC) was approximately 3-fold higher for ceftazidime and over 4-fold higher for avibactam in our patient than in the healthy volunteers. Total clearance was reduced roughly 3-fold to 4-fold for both ceftazidime and avibactam, and the apparent volume of distribution at the steady state (V_ss) was increased 1.5-fold, leading to a half-life (t_1/2) approximately 4 h longer than that observed in healthy volunteers. The plasma pharmacokinetic parameters and dialytic clearance of ceftazidime described in this case are comparable to those previously reported in patients undergoing CVVH (8, 9). There are no published data on the removal of avibactam by CVVH or on its effect on the systemic pharmacokinetics.

The primary pharmacokinetic/pharmacodynamic index of efficacy for β-lactam antibiotics is the time during which serum-free drug concentrations remain above the MIC (fT>MIC) throughout the dosing interval (10). Assuming 10% protein binding of ceftazidime, our patient achieved 100% fT>MIC through the 8-h dosing interval with the dosing regimen of 1.25 g every 8 h while on CVVH. The pharmacodynamic parameter most closely associated with efficacy of the β-lactam inhibitors, including avibactam, appears to be the time during which the free drug concentration is above a threshold concentration required to inhibit β-lactamases throughout the dosing interval. For avibactam, this threshold has been shown to be 1 mg/liter for bactericidal activity and the target proportion of time above that threshold to be 30 to 50% for bactericidal activity (13, 14). Using this endpoint and assuming 7% protein binding, the avibactam concentrations were also above this target for 100% of the dosing interval. The estimated free minimum plasma concentration (C_min) for ceftazidime (28.76 mg/liter) was more than 4-fold above the MIC and was more than 7-fold above the 1 mg/liter threshold for avibactam (7.86 mg/liter) (Table 2). In retrospect, the package insert dosing recommended for patients with an estimated creatinine clearance of 15 to 30 ml/min (0.94 g every 12 h) would have likely also achieved the pharmacokinetic/pharmacodynamic targets for ceftazidime and avibactam, assuming dose-proportional kinetics.

Ultimately, despite achieving target plasma pharmacokinetic/pharmacodynamic endpoints, the patient expired due to persistent infection and multiorgan failure. Blood cultures remained positive for 5 days despite therapy with ceftazidime-avibactam and repeated therapeutic doses of tobramycin. The data obtained here provide clinicians using this agent off-label in critically ill patients on CVVH with some initial dosing guidance. Larger studies exploring multiple patients and dialysis modes are needed to verify the optimal dosing strategy of ceftazidime-avibactam in this patient population. The clearance reported in this case may not be representative of alternate modes and rates of renal replacement therapy or CVVH with postfiltration replacement hemofiltration. The sampling period in this study covered the 8-h dosing interval, but complete elimination phase data are need in order to more accurately quantify the plasma half-life.

MATERIALS AND METHODS

Serial prefiltration blood samples were collected in heparinized K₂ EDTA tubes before, during, at the end, and 4, 6, and 8 h after the start of the fourth infusion of ceftazidime-avibactam. The sampling point at the end of the infusion accounted for the overfill volume and the residual fluid volume in the intravenous tubing. Simultaneous postfiltration and ultrafiltrate samples were obtained at the end of the infusion and at the 6-h time point. Prefiltration blood samples were obtained from an indwelling arterial catheter and postfiltration blood samples from a port on the blood return access line. Pre- and postfiltration blood samples were centrifuged at 2,000 × g for 10 min, and supernatant plasma was frozen at −80°C within 30 min of collection. Ceftazidime and avibactam concentrations were quantified using liquid chromatography-tandem mass spectrometry methods (Keystone Bioanalytical, North Wales, PA). The calibration ranges for ceftazidime and avibactam were 0.04 to 80 mg/liter and 0.01 to 20 mg/liter, respectively, with intra- and interassay error rates of ±10%. (This work involved the use of a marketed drug in the course of medical practice and therefore did not come under the institutional criteria with respect to ethical guidelines for human subject research.)

Steady-state pharmacokinetic parameters for ceftazidime and avibactam were estimated from observed prefiltration plasma concentrations via noncompartmental analysis (WinNonlin Version 7.0, Pharsight Corp., Mountain View, CA). The following parameters were calculated: maximum plasma concentration (C_max), minimum plasma concentration (C_min), half-life (t_1/2), apparent volume of distribution at the steady state (V_ss), total clearance at the steady state (CL_ss), and the area under the concentration-time curve from 0 to 8 h (AUC_0–8). AUC_0–8 was calculated via the linear-up log-down method. Calculations for the estimation of removal of ceftazidime and avibactam by CVVH were as follows (11):

• Extraction ratio (percent): [(prefiltration concentration–postfiltration concentration)/prefiltration concentration] × 100.

• Sieving coefficient (SC): ultrafiltrate concentration/(prefiltration concentration + postfiltration concentration/2).

• CL_CVVH: SC × ultrafiltrate flow rate × correction factor for prefiltration fluid replacement.

• Correction factor: blood flow rate/(blood flow rate + replacement fluid rate) (12).